Project Gutenberg - Characters, Authors and their Gender

The raw data for the study was obtained from a cleaned version of Project Gutenberg, which has 3,036 English books as text files, penned by 142 authors between 1700-1950. Out of the 142 authors, 14 are female.

The first file, a 'books_data_json.txt', contains processed books data is in the form of a JSON (JavaScript Object Notation) text file containing each of the 3036 books information in as many dictionaries. Each book has the title of the book as the main key and another dictionary with the information about the book as its value. The keys in this dictionary are:

‘author_male?’: True if the author is male and False if the author if female
‘author_year’: active decade of the author
‘characters’: a list of all characters in the book
‘character_count’: {‘male’: number of male characters, ‘female’: number of female characters}
‘character_occurrence_count’: {‘male’: number of times male characters are referenced, ‘female’: number of times female characters are referenced}
‘pronoun_count’: {‘male’: number of male pronouns in the book, ‘female’: number of female pronouns in the book}

The second file, 'Authors_Metadata.xlsx', contains information about each author in the dataset, specifically - gender, active year of the author, genres of their work and the number of books by the author in the dataset. All the information about the author was collected manually using resources from the internet like Wikipedia.